;This thesis is based on a theoretical model design that integrates diffraction image output with multi-page data encoding, while fully accounting for the effects of optical blur and noise during the reading process. On the encoding side, we adopt previously developed sparse code tables with different numbers of activated pixels, excluding high-error patterns. This increases the information capacity of each 4×4 sparse code block from 9 bits to as many as 13 bits, while simultaneously reducing errors caused by optical blur. The study is divided into two parts. In the first part, decoding methods are designed for known blur conditions to analyze the joint impact of blur and noise, combined with low-density parity-check (LDPC) coding and decoding. In addition, we propose a boundary-aware decoding method that leverages the blurred boundary pixels of previously decoded blocks as prior information, thereby significantly enhancing the accuracy of blur compensation. In the second part, we further incorporate an iterative probability update mechanism within the LDPC decoding process, applying it to both sparse-coded and non-sparse-coded cases. For the non-sparse-coded case, we also adopt an interleaving storage strategy at the encoding stage to effectively suppress the impact of optical blur on system performance. Experimental results demonstrate that the proposed methods can substantially reduce the bit error rate under various blur and noise conditions, thereby improving the reliability and fault tolerance of holographic data storage systems.
